• 한국측량학회지
• /
• 제40권4호
• /
• pp.315-324
• /
• 2022

일반적인 조정계산에서는 독립변수의 오차는 없다고 가정하고 종속변수의 오차만을 고려하는 최소제곱해를 구한다. 그러나 지상측량에 의해 결정한 3차원 공간좌표나 GNSS (Global Navigation Satellite System) 기반 추정좌표는 성분별로 독립적으로 결정되지 않으므로 모든 성분에 오차가 있을 뿐만 아니라 공분산도 존재한다. 따라서 좌표쌍을 이용한 평면 추정이나 좌표계 변환에서는 모든 성분의 오차를 고려하는 전최소제곱을 적용해야 한다. 이를 위한 다양한 모델이 존재하며, 특별한 제약조건을 제외하면 동등한 해를 제공한다. 본 연구에서는 가우스-헬머트 모델(GHM: Gauss-Helmert Model) 기반 전최소제곱으로 VLBI 타겟이 형성하는 자취를 이용하여 평면의 법선벡터를 추정했으며, 지역좌표계를 세계측지계로 변환하는 계수 결정에도 적용했다. 평면방정식의 경우 기존 최소제곱 방법과 비교해서 법선벡터는 동일하지만 분산요소의 안정성과 타겟 위치에 따른 분산요소 특성을 명확히 확인할 수 있었다. 좌표계 변환계수는 가우스-헬머트 모델을 적용하면 변환 전후 두 좌표계에서 모두 잔차를 계산할 수 있으며, 기존 방식보다 잔차가 더 작아진다.

• Journal of the Korean Data and Information Science Society
• /
• 제20권4호
• /
• pp.749-755
• /
• 2009

Support vector classification (SVC) provides more complete description of the lin-ear and nonlinear relationships between input vectors and classifiers. In this paper. we propose the sparse kernel classifier to solve the optimization problem of classification with a modified hinge loss function and absolute loss function, which provides the efficient computation and the sparsity. We also introduce the generalized cross validation function to select the hyper-parameters which affects the classification performance of the proposed method. Experimental results are then presented which illustrate the performance of the proposed procedure for classification.

• Communications for Statistical Applications and Methods
• /
• 제13권3호
• /
• pp.765-776
• /
• 2006

In this paper we propose an estimation method on the regression model with randomly censored observations of the training data set. The weighted least squares support vector machine regression is applied for the regression function estimation by incorporating the weights assessed upon each observation in the optimization problem. Numerical examples are given to show the performance of the proposed estimation method.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.2480-2484
• /
• 2005

Electrical resistance tomograpy (ERT) maps resistivity values of the soil subsurface and characterizes buried objects. The characterization includes location, size, and resistivity of buried objects. In this paper, truncated least squares (TLS) is presented for the solution of the ERT image reconstruction. Results of numerical experiments in ERT solved by the TLS approach is presented and compared to that obtained by the Gauss-Newton method.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1988년도 한국자동제어학술회의논문집(국제학술편); 한국전력공사연수원, 서울; 21-22 Oct. 1988
• /
• pp.753-758
• /
• 1988

In this paper, we propose a new method of designing a reduced-order controller for a linear discrete-time system. Firstly, we study a design problem for a two-degree-of-freedom control system with a feedforward controller. Secondly, in order to obtain a reduced-order controller, frequency-weighted least squares approximation problems are considered. Thirdly, we propose a synthesis procedure of a reduced-order controller. Finally, an example is given to illustrate the effectiveness of this proposed method.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1990년도 춘계공동학술대회논문집; 한국과학기술원; 28 Apr. 1990
• /
• pp.254-262
• /
• 1990

A predictive functional relationship model is presented for the calibration problem in which the standard as well as the nonstandard measurements are subject to error. For the estimation of the relationship between the two measurements, the ordinary least squares and maximum likelihood estimation methods are considered, while for the prediction of unknown standard measurementswe consider direct and inverse approaches. Relative performances of those calibration procedures are compared in terms of the asymptotic mean square error of prediction.

• 대한전기학회논문지
• /
• 제36권4호
• /
• pp.287-292
• /
• 1987

This paper deals with applications of Haar functions to parameter identification of linear systems. It is first introuduced to a new operational matrix which relates Haar functions and their integrations. The matrix can be used to identify the parameters of unknown linear systems by a least squares estimation. And then, the state equation of given systems is transformed into a computationally convenient algebraic form. This approach provides a more efficient method for the system identification problem.

• 전자공학회논문지A
• /
• 제28A권9호
• /
• pp.671-679
• /
• 1991

In adaptive antenna, several models are known according to a prior knowledge about jammer signal. Among those, Frost model with contraint is generally used however it has the problem that convergence speed is slow and that stability is not good. To improve such problems, this paper proposes constrained NLMS algorithm using LS method. In addition, the result obtained by applying this algorithm to Duvall antenna model is compared with that of Frost model.

• 충청수학회지
• /
• 제24권4호
• /
• pp.895-903
• /
• 2011

A two dimensional version of LSQR iterative algorithm which takes advantages of working solely with the 2-dimensional arrays is developed and applied to the image deblurring problem. The efficiency of the method comparing to the Fourier-based LSQR method and the 2-D version CGLS algorithm methods proposed by Hanson ([4]) is analyzed.

• Journal of applied mathematics & informatics
• /
• 제6권2호
• /
• pp.669-684
• /
• 1999

We are interested in the problem of fitting a curve to a set of points in the plane in such a way that the sum of the squares of the orthogonal distances to given data points ins minimized. In[1] the prob-lem of fitting circles and ellipses was considered and numerically solved with general purpose methods. Especially in [2] H. Spath proposed a special purpose algorithm (Spath's ODF) for parabolas y-b=$c($\chi$-a)^2$ and for rotated ones. In this paper we present another parabola fitting algorithm which is slightly different from Spath's ODF. Our algorithm is mainly based on the steepest descent provedure with the view of en-suring the convergence of the corresponding quadratic function Q(u) to a local minimum. Numerical examples are given.